Empathy List Archives

WG

William Grissom

Mon, Jun 16, 2014 4:47 PM

Hi all,

I am using a USRP1 with BasicTX and BasicRX boards and GNU Radio for basic pulse-and-acquire NMR experiments at 20 MHz. Overall everything works as expected except that there is a small frequency offset between transmit and receive, which I have measured to be on the order of a few Hz. It seems to change each time I power on the radio. I have written code to estimate it and post-process the data to remove it prior to averaging measurements, but I would obviously prefer if it didn't exist in the first place. My flowgraph comprises only a block pulse generator and USRP Sink for transmit, and a USRP Source and scope for receive. I am using a 250 kHz sample rate, and the USRP blocks use the same center frequency variable.

I am wondering if this is an issue that I can fix with a simple change in my flowgraph settings, or is there something hardware-related to it?

Thanks very much for any help,
Best,
Will Grissom

Hi all, I am using a USRP1 with BasicTX and BasicRX boards and GNU Radio for basic pulse-and-acquire NMR experiments at 20 MHz. Overall everything works as expected except that there is a small frequency offset between transmit and receive, which I have measured to be on the order of a few Hz. It seems to change each time I power on the radio. I have written code to estimate it and post-process the data to remove it prior to averaging measurements, but I would obviously prefer if it didn't exist in the first place. My flowgraph comprises only a block pulse generator and USRP Sink for transmit, and a USRP Source and scope for receive. I am using a 250 kHz sample rate, and the USRP blocks use the same center frequency variable. I am wondering if this is an issue that I can fix with a simple change in my flowgraph settings, or is there something hardware-related to it? Thanks very much for any help, Best, Will Grissom

MM

Marcus Müller

Mon, Jun 16, 2014 5:52 PM

Hello Will,

the Basic daughterboards don't have any mixers, so the only oscillator
involved in BasicTX-RX transmission are these of the DAC/ADC. Assuming
these are on the same USRP1, there should not be any offset, since both
daughterboards are driven off the same LO.
If they are on two different USRP1, then it's physically impossible to
have both oscillators on exactly the same frequency, which would explain
some offset even with the rather high quality oscillators on USRPs.

Greetings,
Marcus

On 16.06.2014 18:47, William Grissom via USRP-users wrote:

Hi all,

I am using a USRP1 with BasicTX and BasicRX boards and GNU Radio for basic pulse-and-acquire NMR experiments at 20 MHz. Overall everything works as expected except that there is a small frequency offset between transmit and receive, which I have measured to be on the order of a few Hz. It seems to change each time I power on the radio. I have written code to estimate it and post-process the data to remove it prior to averaging measurements, but I would obviously prefer if it didn't exist in the first place. My flowgraph comprises only a block pulse generator and USRP Sink for transmit, and a USRP Source and scope for receive. I am using a 250 kHz sample rate, and the USRP blocks use the same center frequency variable.

I am wondering if this is an issue that I can fix with a simple change in my flowgraph settings, or is there something hardware-related to it?

Thanks very much for any help,
Best,
Will Grissom

USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Hello Will, the Basic daughterboards don't have any mixers, so the only oscillator involved in BasicTX-RX transmission are these of the DAC/ADC. Assuming these are on the same USRP1, there should not be any offset, since both daughterboards are driven off the same LO. If they are on two different USRP1, then it's physically impossible to have both oscillators on exactly the same frequency, which would explain some offset even with the rather high quality oscillators on USRPs. Greetings, Marcus On 16.06.2014 18:47, William Grissom via USRP-users wrote: > Hi all, > > I am using a USRP1 with BasicTX and BasicRX boards and GNU Radio for basic pulse-and-acquire NMR experiments at 20 MHz. Overall everything works as expected except that there is a small frequency offset between transmit and receive, which I have measured to be on the order of a few Hz. It seems to change each time I power on the radio. I have written code to estimate it and post-process the data to remove it prior to averaging measurements, but I would obviously prefer if it didn't exist in the first place. My flowgraph comprises only a block pulse generator and USRP Sink for transmit, and a USRP Source and scope for receive. I am using a 250 kHz sample rate, and the USRP blocks use the same center frequency variable. > > I am wondering if this is an issue that I can fix with a simple change in my flowgraph settings, or is there something hardware-related to it? > > Thanks very much for any help, > Best, > Will Grissom > > > > > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

MD

Marcus D. Leech

Mon, Jun 16, 2014 5:56 PM

On 06/16/2014 12:47 PM, William Grissom via USRP-users wrote:

Hi all,

I am using a USRP1 with BasicTX and BasicRX boards and GNU Radio for basic pulse-and-acquire NMR experiments at 20 MHz. Overall everything works as expected except that there is a small frequency offset between transmit and receive, which I have measured to be on the order of a few Hz. It seems to change each time I power on the radio. I have written code to estimate it and post-process the data to remove it prior to averaging measurements, but I would obviously prefer if it didn't exist in the first place. My flowgraph comprises only a block pulse generator and USRP Sink for transmit, and a USRP Source and scope for receive. I am using a 250 kHz sample rate, and the USRP blocks use the same center frequency variable.

I am wondering if this is an issue that I can fix with a simple change in my flowgraph settings, or is there something hardware-related to it?

Thanks very much for any help,
Best,
Will Grissom

USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

There should be zero frequency difference between RX and TX in this
case, since everything is driven off a common clock, and you have no
analog up/downconversion happening.

Simplify, for purposes of experiment. Issue a CW tone on the TX side,
and see what the RX sees.

--
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org

On 06/16/2014 12:47 PM, William Grissom via USRP-users wrote: > Hi all, > > I am using a USRP1 with BasicTX and BasicRX boards and GNU Radio for basic pulse-and-acquire NMR experiments at 20 MHz. Overall everything works as expected except that there is a small frequency offset between transmit and receive, which I have measured to be on the order of a few Hz. It seems to change each time I power on the radio. I have written code to estimate it and post-process the data to remove it prior to averaging measurements, but I would obviously prefer if it didn't exist in the first place. My flowgraph comprises only a block pulse generator and USRP Sink for transmit, and a USRP Source and scope for receive. I am using a 250 kHz sample rate, and the USRP blocks use the same center frequency variable. > > I am wondering if this is an issue that I can fix with a simple change in my flowgraph settings, or is there something hardware-related to it? > > Thanks very much for any help, > Best, > Will Grissom > > > > > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > There should be *zero* frequency difference between RX and TX in this case, since everything is driven off a common clock, and you have no analog up/downconversion happening. Simplify, for purposes of experiment. Issue a CW tone on the TX side, and see what the RX sees. -- Marcus Leech Principal Investigator Shirleys Bay Radio Astronomy Consortium http://www.sbrac.org

usrp-users@lists.ettus.com

USRP1 frequency difference between TX+RX